Research window: Past 24 hours (2026-09-06 11:18 ~ 2026-09-07 11:18 Beijing Time) Sources: RT-Thread official news API, GitHub (org: RT-Thread / main repo commits+PRs / review queue / topic:rt-thread ecosystem repos), Google News aggregation (zh-CN / en-US, via proxy), Chinese industry media including Electronic Engineering Album (see Appendix: Source Verification Table for details)

Index

  • I. Kernel and Release Updates
    • End of three-day merge drought: Monday morning batch merges 4 PRs, covering mmcsd AC5 compatibility, MPU Kconfig, eclipse toolchain, hpmicro C++ build (09-07)
    • Review queue activity concentrated: #11774 re-pushed same day after rejection, #11775 forwarded for re-review, #11770 receives inline suggestions; new PR #11781 lands (09-07)
  • II. Official and Community Events
    • Official news API sees no additions for the eighteenth consecutive day, elexcon 2026 opens in 2 days (09-07)
  • III. Ecosystem and Vendor Porting
    • canopennode-rtt pauses during window, author wdfk-prog’s official eclipse toolchain PR merges same day — ecosystem advances on two fronts (09-07)
    • No other activity in the topic:rt-thread ecosystem during the window (09-07)
  • IV. Industry Applications
    • No new entries (no new industry case reports during the window)
  • V. Summary and Trend Observations
    • Three-day trend (09-05 ~ 09-07)
  • Appendix: Source Verification Table

I. Kernel and Version Dynamics

1.1 Three-Day Zero-Merge Streak Ends: Monday Morning Batch Merges 4 PRs Covering mmcsd AC5 Compatibility, MPU Kconfig, eclipse Toolchain, and hpmicro C++ Build

Date: 2026-09-07 (merge window 10:19–11:18 Beijing time, i.e., 02:19:23Z–03:18:42Z)　Source: RT-Thread/rt-thread commits (PR #11690, PR #11766, PR #11762, PR #11777)

The main rt-thread repository ended three consecutive windows (three days) of zero merges since #11776 on 09-03, merging 4 PRs in a single hour on Monday morning:

  • #11690 [device][mmcsd] Fix Arm Compiler 5 build compatibility (submitted by CYFS3, created 08-10, the longest-queued at roughly 4 weeks): systematically fixes AC5 compilation compatibility issues in MMC/SD and block device code—the generic SWITCH command name conflicts with certain chip register macros (renamed to MMC_SWITCH with corresponding updates to Phytium SDIF call sites); provides named SD status field access for ARMCC5 (other compilers continue using the anonymous field interface); replaces block device list traversal that relies on typeof with explicit rt_list_entry; adds partition structure definitions when DFS is disabled; and cleans up void * pointer arithmetic.
  • #11766 [mprotect][kconfig] Add safe defaults for MPU region count options (submitted by RogerNext, created 08-28): adds Kconfig defaults for NUM_MEM_REGIONS / NUM_CONFIGURABLE_REGIONS / NUM_EXCLUSIVE_REGIONS. Previously, when RT_USING_HW_STACK_GUARD was enabled without explicit configuration, rtconfig.h generated empty macros (#define NUM_CONFIGURABLE_REGIONS), which expanded into invalid C expressions ((2 + ), for (i = 0U; i < ; i++)), causing MPU code compilation failures on cortex-m4/m7/m33.
  • #11762 [tools][eclipse] Sync assembler include paths (submitted by wdfk-prog, created 08-28): syncs assembler include paths in the eclipse plugin project, a build consistency fix for the official toolchain (eclipse plugin).
  • #11777 [bsp][hpmicro] wire CXX/CXXFLAGS from rtconfig into scons Environment (submitted by kaidegit, created 09-01): when the hpmicro BSP (HPM series from HPMicro) enables RT_USING_CPLUSPLUS, the scons Environment did not wire in CXX/CXXFLAGS from rtconfig and fell back to the host g++, causing C++ group compilation flags to be treated as a single unquoted argument and breaking the build; the fix injects CXX/CXXFLAGS into the scons environment.

This batch coincided with maintainer Rbb666’s intensive review activity the same morning (see 1.2), shifting the main repository from an “accumulation phase” to “dual activity in merges + reviews”; the release remains v5.2.2 (2025-10-31), with no new version release action yet.

1.2 Review Queue Intensifies: #11774 Re-pushed Same Day After Rejection, #11775 Reassigned for Re-review, #11770 Receives Inline Suggestion; New PR #11781 Lands

Date: 2026-09-07 (Beijing time 10:16–11:18)　Source: RT-Thread/rt-thread pulls

During the same period as the merge batch, maintainer Rbb666 handled the review queue intensively:

  • #11774 [components][sal] Fix SIOCGIFCONF buffer overflow (Huoyanlifusu): received Rbb666’s CHANGES_REQUESTED review at 02:30:45Z; the author pushed a new commit the same day at 03:04:48Z ([components][sal] Fix SIOCGIFCONF buffer overflow) in response—a fast turnaround.
  • #11775 [bsp][c908] rt-smart and MMU Sv39/48 support (QEMU xiaohui c908, including OpenSBI SMP) (BadmintonMaster): Rbb666 commented at 02:18:28Z, “@rcitach please help review,” reassigning to another maintainer for re-review; this PR provides rt-smart and OpenSBI SMP support for DAMO XuanTie C908 on the QEMU xiaohui platform, RISC-V related.
  • #11770 fix[usb]: stop aliasing rt_ringbuffer as rt_serial_rx_fifo in VCOM (Finder16): Rbb666 left an inline suggestion at 03:17:12Z on components/legacy/usb/usbdevice/class/cdc_vcom.c:358—adding a polling receive path for rx_bufsz == 0 (when Serial V2 sets config.rx_bufsz to 0 before opening the device and uses RT_SERIAL_RX_BLOCKING, it enters polling mode and does not allocate serial_rx), along with an example patch.
  • #11772 [components][asan] Add runtime AddressSanitizer for heap (RogerNext): received Rbb666’s review comment (COMMENTED) at 02:16:58Z.

On the new PR front: #11781 [BSP][STM32] Adapt Ethernet driver to new HAL API (CYFS3, created 09-06 15:17 Beijing time) landed within the window, adapting the STM32 Ethernet driver to the new HAL API—part of the same STM32 driver modernization thread as #11689 (stm32 SPI BDMA, moment-NEW) in the queue. No new issues within the window.

II. Official and Community Activities

2.1 Official News API Silent for the 18th Consecutive Day; elexcon 2026 Opens in 2 Days

Date: 2026-09-07　Source: RT-Thread Official News API

The official news API (rt_org category) still shows 5 old entries, the latest still being the 2026-06-17 Abu Dhabi investment forum report—the 18th consecutive day with no additions. elexcon 2026 in Shenzhen (September 9–11, with RT-Thread exhibiting; the “AI Hardware Full-Stack” preview was recorded on 08-30) opens in 2 days, with Wednesday this week being the next external window. On the Chinese media side (Google News via proxy checks for “RT-Thread” when:3d, 睿赛德 when:7d, 睿擎派 OR 睿擎 when:7d) and the English side (“RT-Thread” when:2d), there were zero new reports within the window; the latest media content remains the 09-04 Electronic Engineering Album community work share (recorded yesterday), and the pre-event silence continues.

III. Ecosystem and Vendor Adaptation

3.1 canopennode-rtt Pauses Within the Window; Author wdfk-prog’s Official eclipse Toolchain PR Merged the Same Day—Ecosystem Advances on Two Fronts

Date: 2026-09-07　Source: wdfk-prog/canopennode-rtt, RT-Thread/rt-thread PR #11762

The external branch repository canopennode-rtt has had no new commits within the window since 09-06 10:06 (CiA401 analog event and fail-safe support, recorded yesterday), with the CiA402/CiA401 dual-profile expansion entering a phase pause. Notably, the same developer wdfk-prog’s contribution to the official repository landed today: their eclipse toolchain PR #11762, submitted on 08-28, was merged into the main repository this morning (see 1.1), while their led CAN framework refactor #11779 remains under review (last comment from CYFS3 on 09-03). The dual-track momentum of “external CANopenNode ecosystem port (canopennode-rtt) + official CAN framework/toolchain contributions (#11779/#11762)” remains unchanged, and Monday’s merge window marks the first same-day landing of achievements from both tracks.

3.2 No Other Activity in the topic:rt-thread Ecosystem Within the Window

Date: 2026-09-07　Source: GitHub topic:rt-thread

topic:rt-thread ecosystem check: no new repository pushes since 09-06 (search topic:rt-thread pushed:>2026-09-06 returned 0 hits); the packages index repository still has no updates (last was the 09-01 CANopenNode index, already recorded); high-frequency repositories such as zlgopen/awtk had no commits directly related to RT-Thread within the window. Ecosystem focus remains on canopennode-rtt and its author’s main repository contributions.

IV. Industry Applications

No new entries. No new industry application/deployment case reports within the window; the industry side has remained silent for the past three days. The most recently recorded industry content is the 08-27 Malaysia urban IoT case and the 08-24 RISC-V + AI dual-track summit report (Electronic Engineering Album). RT-Thread’s latest external developments in embodied intelligence (Lingang Dishui Zhihui) and industrial (Ruiqing industrial platform) directions are expected to be released in a concentrated manner with the opening of elexcon 2026 on 09-09.

V. Summary and Trend Observations

  • Kernel side: The three-day zero-merge streak ended, with a batch of 4 PRs merged within one hour on Monday morning 10:19–11:18—the mmcsd AC5 compatibility fix (including synchronized Phytium SDIF call sites, the longest-queued #11690 at roughly 4 weeks), MPU region count Kconfig defaults (fixing cortex-m4/m7/m33 compilation failures), eclipse toolchain assembler include sync, and hpmicro BSP C++ build fix; during the same period, maintainer Rbb666 conducted intensive reviews—#11774 re-pushed by the author the same day after rejection, #11775 reassigned to rcitach for re-review, #11770 received an inline patch suggestion, #11772 received a review comment; new PR #11781 (STM32 Ethernet HAL adaptation) landed, with zero new issues in the window and the release still at v5.2.2. The main repository has shifted from accumulation to “dual activity in merges + reviews,” and given the imminent opening of elexcon 2026, the intent to clear the review queue before the event is evident.
  • Ecosystem side: canopennode-rtt paused within the window (last commit on 09-06 morning, already recorded), but its author wdfk-prog’s official eclipse toolchain PR #11762 landed the same day as the external port, advancing on both “external ecosystem + official contributions” tracks; the rest of the topic ecosystem showed no activity.
  • Official/media side: The official news API has been silent for the 18th consecutive day; the media side saw zero additions in the 24h window (proxy and source checks normal); elexcon 2026 (09-09–11, Shenzhen) is 2 days away, and pre-event silence is the expected rhythm.
  • Three-day trend (09-05 to 09-07): Official silence day 16, env merges WebUI keep_alive (#294), sdio CIS leak PR #11780, CiA402 diagnostics/EMCY/controller side and CiA401 branch launch (09-05) → official silence day 17, CiA401 analog event and fail-safe land, gdr FreeRTOS adaptation wraps up (09-06) → official silence day 18, main repository 4-PR batch merge and intensive review queue activity, new PR #11781 STM32 Ethernet adaptation (09-07). The main advancement thread has shifted from “official toolchain WebUI iteration + CANopenNode dual-profile expansion” to “dual activity in main repository merges/reviews (primarily toolchain and BSP build fixes) + pre-event queue cleanup,” with external activities building up for elexcon 2026.

Appendix: Source Verification Table

Source Verification Result
RT-Thread Official News API rt_org category still shows 5 old entries, latest 06-17 Abu Dhabi report, 18th consecutive day with no additions
GitHub org: RT-Thread Pushes within 7 days: rt-thread (09-07T03:18Z, this morning’s merge batch), env (09-04, #294 recorded), packages (09-01); org activity within the window concentrated in the main repository
rt-thread main repository commits 4 merges within the window (09-07T02:19Z–03:18Z: #11777/#11766/#11762/#11690, see 1.1), previous merge 09-03 #11776
rt-thread new PRs / issues 1 new PR #11781 (STM32 Ethernet HAL adaptation, created 09-06T07:17Z); 0 new issues within the window
rt-thread review queue #11774 re-pushed by author same day after CHANGES_REQUESTED (03:04Z); #11775 reassigned by Rbb666 to rcitach for re-review; #11770 Rbb666 inline suggestion (cdc_vcom.c rx_bufsz==0 polling path); #11772 Rbb666 review comment; #11779 CAN refactor still open (no new progress)
rt-thread releases No new release (latest v5.2.2, 2025-10-31)
packages commits No commits within the window (last 09-01 CANopenNode index, recorded)
topic:rt-thread ecosystem repositories No new repository pushes since 09-06 (search returned 0 hits); canopennode-rtt zero commits (last 09-06T02:06Z, recorded); awtk etc. no topic-related commits
Google News (“RT-Thread” when:3d / 睿赛德 when:7d / 睿擎派 OR 睿擎 when:7d, zh) Only hit the 09-04 Electronic Engineering Album old article (recorded); 睿赛德/睿擎派 0 hits; bare queries normal, proxy and parsing no anomalies
Google News (“RT-Thread” when:2d, en) 0 hits
Electronic Engineering Album Direct site access unreachable (timeout); entries double-verified via gnews titles + Tavily index summaries; no new RT-Thread reports within the window
Hacker News fetch returned 2 hits, both noise (ThreadSanitizer review, McDonald’s incident thread mismatch), excluded